Improved understanding of the potential for ventricular-induced lung damage has challenged the conventional approach to mechanical ventilation and inspired a number of recent therapeutic innovations whose value remains unproven. In the conventional approach to acute lung injury (ALI), large tidal volumes and rapid cycling frequencies are used to 'normalize' pH and/or PaCO2. As a rule, airway pressures are monitored but not rigidly constrained. Recent studies demonstrating the heterogeneity of damage in ALI suggest, however, that such an approach puts functioning portions of the injured lung at risk for further damage. Undamaged areas with preserved compliance are subjected to local hyperventilation, over distention, and inhibition of surfactant. High shear stresses may result from rapid inflation to elevated transalveolar pressures, especially at the junctions of structures that are mobile (e.g., aerated alveoli) and immobile (e.g., consolidated alveoli, or con ducting airways). Failure to maintain a certain minimum lung volume in the setting of ALI or excessive tidal excursions may also produce or accentuate lung damage. Limiting peak transalveolar pressure, minimizing tidal excursions of alveolar pressure, reducing the shearing stresses applied to the lung parenchyma, and preserving a minimum end-expiratory alveolar volume may each help prevent ventilator-induced lung dysfunction. However, extended inspiratory time fractions and CO2, retention may be inevitable and poorly tolerated consequences of such a lung-protecting, pressure-limiting ventilatory strategy. Hypercapnia may be attenuated by the adjunctive use of tracheal gas insufflation. The overall objective of this proposed series of projects is to use biophysical principles and empirical data from the laboratory and bedside to define an optimal clinical approach for the ventilatory support of patients with ALI. This work will proceed along three intersecting lines in experimental animal models and patients with ALI: mathematical modelling of lung mechanics, gas exchange and ventilatory methods; experimental testing of alternative methods of ventilatory support in animal surrogates of ALI; and clinical trials of pressure targeted ventilation. We specifically intend to define the biomechanical, pathological, and functional consequences of specific decisions made by the clinician relative to choosing ventilator mode and settings, and to develop tracheal gas insufflation as an adjunctive technique to help accomplish effective gas exchange without extending tissue injury.